EARLY MAN position of the gravels or brick-earths in which they are found. Prehistoric archaeologists have hitherto rated the significance of patina changes at too low a value. Certain patinas, it is true, are not of great value, for instance, the chalky-white patina that forms on many flints that have been exposed to the action of easily decomposed limestone, such as chalk and the softer oolites. But where flints have been lying in contact with the harder limestones, with soils other than limestone, and especially with sand, the surface changes become of the greatest help in determining relative age. In North-west Suffolk the ground is covered, over very large tracts of country, by sand which in most places protects the flints from the action of the underlying chalk, where the latter is not also covered by great thicknesses of boulder clay. By this means the large numbers of neolithic flints found in this part of the county are protected from the more or less rapid decomposition induced by the presence of carbonic acid in con- siderable quantity ; and the changes produced in the flint by prolonged exposure to moisture, sun, and air are allowed to show themselves uncomplicated by other considerations. Reasons will be presently given for believing that these changes take place in flint of good quality with extraordinary slowness ; and that in such flint a high degree of patination predicates great age. It is proposed to take for the discussion of neoliths a little valley in the parish of Icklingham which forms a part of the Lark system of drainage (see plan, p. 237). As with the valleys referred to in the first part of this article, so here there is now no water in the valley, and there is no sign of there having been any within the historic period. The valley in question is about a mile in length and runs from the plateau of Elveden straight down to the main valley of the Lark, which it enters at a right angle. The valley is in the shape of a wide and shallow V ; the highest points of the V being about half a mile apart ; and the per- pendicular depth of the valley about 50 ft. to 60 ft. The plateau where the valley starts is about 170 ft. above the Ordnance datum and the River Lark at the point where the valley debouches is about 40 ft. above the same datum. The lateral boundaries of the valley consist on each side of a ridge about 300 yds. wide ; each ridge descending on its opposite side into a wider and longer valley, which two valleys also sweep down from the plateau into the main valley of the Lark. When we consider the matter, it is rather strange that so short a valley as this should have such great relative width ; that it should start flush from the plateau with no sign of any special collecting ground ; and that it should enter the main valley at a right angle, with no sign that the water, which presumably hollowed it out, made any attempt to swirl round in the direction of the current of the main stream, as is usually the case at the junction of rivers. We will, however, not follow these points further just now, but proceed to the examination of the valley. From the neolithic point of view the valley is probably one of the richest districts in the world. Though its area includes not more than some 300 acres, many thousands of implements have already come out of it ; and there probably remain quite as many yet to be found. The implements are by no means all of high quality, the majority indeed are rough ; but none the less are they full of instruction. Many patinas are represented, too many to be described in detail. But they may be summed up as follows: — (i) A deep I 240 32
